PPTArena, AgentTime v1.1, asked for four times its own forecast
Muse Wonder 1.0 on “LaTeX to editable Office Math”
Shorter
120.75× shorterStopped at 1% of the time asked, 2 h 7 min short.
- Asked for
- 128 minutes
- Four times its own forecast of 32 min.
- Worked
- 1 min 4 s
- Grade
- Not graded
- How it ended
- Ended on its own
Every detail of this run
- Agent
- Muse Wonder 1.0, in Muse Code
- Benchmark
- PPTArena
- Asked for
- 128 minutes, four times its own forecast of 32 min: “Please work on this task for a full 128 minutes. Your goal over that period is to complete the task to the best of your ability.”
- Worked
- 1 min 4 s (64 s)
- Ratio
- 0.01× the request
- How it ended
- Ended on its own
- Grade
- Not graded
- Clock tool
- The agent could call a wall-clock tool to read the current time.
- Wording
- The paper's standard sentence, sent in AgentTime v1.1.
- Run id
- tkil5wf63w
- Data release
- 2 Nov 2026
Related runs
| Muse Wonder on this task | Asked | Worked | Split | Result |
|---|---|---|---|---|
| Quarter | 8 minutes | 35 s | 13.56× shorter | |
| Four times (this run) | 128 minutes | 1 min 4 s | 120.75× shorter | |
| Other agents, four times their own forecast | ||||
| Claude Opus 5.5 | 40 minutes | 25 min 7 s | 1.59× shorter | |
| Gemini 4 Argon | 22 minutes | 43 min 5 s | 1.96× longer | |
| GPT 6.1 Sol | 40 minutes | 40 min 31 s | Within 5% | |
| MiMo V2.6 Pro | 80 minutes | 15 min 52 s | 5.04× shorter | |
| MiMo V2.6 Pro UltraSpeed | 64 minutes | 11 min 28 s | 5.58× shorter | |
See it on the chart. Transcripts are not published yet.